Clustering Based on MST A minimum spanning tree of a weighted graph can be used for clustering if the tree is acyclic and each edge of the tree represents a bridge. Thus, deleting an edge from a graph G's MST separates G into two clusters, and continuing in this manner yields clusters. To obtain k clusters, we must eliminate the k 1 heaviest edges from the graph's MST. Because it is difficult to guess the value of k, a threshold may be used to remove from the MST all edges with a weight greater than to produce the clusters. This algorithm's further phases can now be defined. 1. Input: A weighted graph G = (V, E, w), threshold τ . 2. Output: C = {C1,C2,...,Ck }, k clusters of G. 3. Find MST T of G. 4. Remove all edges from G that have a weight larger than τ . 5. Find the components of GN which are the clusters of G Make Python Implementation
Clustering Based on MST
A minimum spanning tree of a weighted graph can be used for clustering if the tree is acyclic and each edge of the tree represents a bridge. Thus, deleting an edge from a graph G's MST separates G into two clusters, and continuing in this manner yields clusters. To obtain k clusters, we must eliminate the k 1 heaviest edges from the graph's MST. Because it is difficult to guess the value of k, a threshold may be used to remove from the MST all edges with a weight greater than to produce the clusters. This
1. Input: A weighted graph G = (V, E, w), threshold τ .
2. Output: C = {C1,C2,...,Ck }, k clusters of G.
3. Find MST T of G.
4. Remove all edges from G that have a weight larger than τ .
5. Find the components of GN which are the clusters of G
Make Python Implementation
Step by step
Solved in 3 steps with 1 images
